Menu Content/Inhalt
Seminars Print
previous year previous month next month next year
See by year See by month See by week See Today Search Jump to month
James Mitchell, NIESR Print
Thursday, 08 October 2009, 12:15 - 13:15

Measuring Output Gap Uncertainty

Abstract: Although central to monetary policy in practice, output gap measurements are contaminated by model uncertainty. In this paper, we propose a methodology for producing predictive densities for the output gap in real time using a large number of vector autoregessions in inflation and different measures of the output gap. Density combination via a linear mixture of experts framework produces (potentially non-Gaussian) ensemble densities for the unobserved output gap. In our application, we show that data revisions contribute substantially to US output gap uncertainty using a variety of output gap measurements derived from various detrending filters. The resulting VAR ensemble produces well-calibrated forecast densities for US inflation in real time, in contrast to those from simple univariate autoregressions which ignore the contribution of the output gap. Broadening our empirical analysis to consider output gap measures derived from linear time trends generates very different point estimates of the output gap, yet the inflation forecast densities are well-calibrated. Combining evidence from both types of output gap measurements indicates strong multi-modality in the predictives for the unobserved output gap. The twin peaks associated with the two detrending methodologies often point to output gaps of opposite sign, reflecting the pervasive nature of model uncertainty.

James Mitchell, NIESR

Location: S 12.227
Contact: Claude Adan, This e-mail address is being protected from spam bots, you need JavaScript enabled to view it